Loss is applied to all assistant turns in the multi-turn trajectory, enabling the model to learn environment observation, action selection, tool use, and recovery from errors.

Training Configuration

Base model: Qwen/Qwen3-4B-Instruct-2507
Method: LoRA (full precision base)
Max sequence length: 2048
LoRA: r=64, alpha=128
Phase 1 dbbench_sft_dataset_react_v4 500 samples
Epochs: 1
Learning rate: 2e-6
Phase 2 sft_alfworld_trajectory_dataset_v5
Epochs: 4
Learning rate: 2e-6 * 0.35
gradient_accumulation_steps=8
lr_scheduler_type="cosine"
warmup_ratio=0.25

Usage

python

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

merged_model_path = "today55/lora-repo-Adv4"

tokenizer = AutoTokenizer.from_pretrained(merged_model_path)
model = AutoModelForCausalLM.from_pretrained(
    merged_model_path,
    torch_dtype=torch.float16,
    device_map="auto"
)

Sources & Terms (IMPORTANT)

Training data: u-10bei/dbbench_sft_dataset_react_v4,u-10bei/sft_alfworld_trajectory_dataset_v5

Dataset License: MIT License. This dataset is used and distributed under the terms of the MIT License. Compliance: Users must comply with the MIT license (including copyright notice) and the base model's original terms of use.